In recent years, the through-wall imaging (TWI) systems have garnered tremendous interest due to their potential applications in surveillance, search–rescue operations, and security monitoring, enabling the detection and visualization of objects behind opaque obstacles such as walls. The multiple-input multiple-output (MIMO) TWI system has been developed to detect hidden targets behind walls. MIMO-based TWI considerably reduces the time it takes to scan the walls. This study presents an experimental evaluation of a MIMO-based TWI system using a dataset of targets with varying dielectric properties placed at uniform and nonuniform distances behind a wall. The B-scan data are collected for different types of targets for contrast target detection. To effectively suppress clutter and enhance target detectability, we employed a range of advanced signal processing techniques, including singular value decomposition (SVD), SVD-based noise subspace exploitation, the optimum shrinkage algorithm, and background subtraction using data-driven thresholding. The target-to-clutter ratio of the final image was computed to evaluate the effectiveness of each clutter removal technique and to facilitate a comparative analysis of the results. The imaging results demonstrated that the system could detect single and multiple targets, including those composed of different materials.